Ankur Pandey

Computer Vision & Machine Learning Engineer with hands-on experience training and deploying YOLO-based object detection models, achieving ~90% mAP on retail shelf analytics. Skilled in Python, OpenCV, PyTorch, Ultralytics YOLO, and Flask/FastAPI REST API deployment, with a track record of shipping production-ready inference pipelines under 300ms latency.

Junior Machine Learning Engineer at Love in Store Technologies Pvt. Ltd. (2026-04 – Present)

• Improved retail shelf product recognition to ~90% mAP by fine-tuning a YOLOv8 model via transfer learning on a custom-annotated 5,000+ image dataset (Roboflow, PyTorch, OpenCV)
• Cut inference response time to under 300ms per image by building and optimizing a Flask/FastAPI REST endpoint for real-time model serving
• Reduced integration effort for 2 downstream product teams by delivering a documented, production-ready inference pipeline with JSON output contracts
• Increased annotation throughput and label quality by standardizing a QC process across train/validation/test dataset splits

Web Developer Intern at Humrobo Technologies (2023-07 – 2024-01)

• Delivered 6+ pixel-perfect responsive pages by translating Figma designs into HTML/CSS/JS with a mobile-first, media-query-driven layout
• Eliminated cross-browser rendering bugs across Chrome/Safari/Firefox by auditing and refactoring CSS, improving UI accessibility

Bachelor of Technology in Computer Science and Engineering – Noida Institute of Engineering and Technology (2022 – 2026)

Industry-Ready Certification in Full-Stack Development – NxtWave Disruptive Technologies (2022-11 – 2026)